European Journal of Heart Failure (2014) 16, 729–736 doi:10.1002/ejhf.105

Burden of new hospitalization for heart failure: a population-based investigation from Italy Giovanni Corrao1*, Arianna Ghirardi1, Buthaina Ibrahim1, Luca Merlino2, and Aldo Pietro Maggioni3 1 Dipartimento

di Statistica e Metodi Quantitativi, Sezione di Biostatistica, Epidemiologia e Sanità Pubblica, Università di Milano-Bicocca, Milan, Italy; 2 Direzione Generale Salute, Unità Organizzativa Governo dei dati, delle strategie e piani del sistema sanitario, Regione Lombardia, Milan, Italy; and 3 Associazione Nazionale Medici Cardiologi Ospedalieri Research Center, Florence, Italy Received 20 December 2013; revised 9 April 2014; accepted 11 April 2014 ; online publish-ahead-of-print 7 May 2014

Aims

Heart failure has been described as one of the emerging pandemics of the 21st century. This report aims to measure the burden of new hospitalization for heart failure in the population of an Italian region of nearly 10 million inhabitants. ..................................................................................................................................................................... Methods Data were retrieved from healthcare utilization databases covering the population of the Italian region of Lombardy. and results We identified patients who were hospitalized for the first time with a primary diagnosis of heart failure (hospitalized heart failure, HHF) during 2011. Incident HHF cases were used for measuring incidence rates and exploring mortality, re-hospitalizations, and healthcare costs on the 1-year time horizon after the index hospitalization. Out-of-hospital mortality, hospitalizations, and healthcare costs were also measured in a referent cohort free from heart failure hospitalization and matched 1:1 by gender and age with the HHF cohort. The overall HHF incidence rate was 32 and 20 events per 10 000 person-years in men and women, respectively. The incidence increased steeply with age in both genders. Among newly hospitalized patients, 7% died during hospitalization. Among survivors, cumulative out-of-hospital mortality and hospital readmission were 24% and 59%, respectively. The average per capita cost was €11 000, the main cost being hospitalizations. Mortality, readmissions, and costs experienced by HHF patients of 88, 75, and 79%, respectively, exceeded those of the referent cohort. ..................................................................................................................................................................... Conclusions The main burden associated with HHF is related to hospitalizations. Effective treatment options that decrease hospitalization rates could reduce patients’ suffering and offer considerable cost savings.

.......................................................................................................... Healthcare costs • Outcomes

Healthcare utilization database •

Introduction The rate of heart failure (HF) hospitalization, particularly in developed countries with ageing populations, has increased progressively over the past decade,1 – 3 making HF the most common reason for hospital admission in elderly patients.4 A diagnosis of HF carries substantial risk of morbidity and mortality. Patients with hospitalization for HF (HHF) continue to have mortality and readmission rates approaching 15% and 30%, respectively, within 30–60 days post-discharge.5,6 The expenditure for direct medical costs (expressed as billion euro) was 53 in the USA in 2010,7 2.7 in

.................................

Keywords

Heart failure •

Incidence •

Mortality •

France in 1990,8 and 0.76 in Poland in 2011.9 All these data, taken together, explain why HF has become a major public health concern, and has been described as one of the ‘emerging’ pandemics for the 21st century.10 Hospital-based registries, sometimes including cardiology wards on a national basis,6,11 report data describing clinical features and outcomes of HHF patients; however, medium- to long-term follow-up is generally not available.12 In contrast, there is a paucity of data regarding the magnitude and outcomes associated with HF from the more generalizable perspective of a population-based investigation.13 The purpose of the present study is to measure

*Corresponding author. Dipartimento di Statistica e Metodi Quantitativi, Sezione di Biostatistica, Epidemiologia e Sanità Pubblica, Università degli Studi di Milano-Bicocca, Via Bicocca degli Arcimboldi, 8, Edificio U7, 20126 Milano, Italy. Tel: +39 02 64485854, Fax: +39 02 64485899, Email: [email protected]

© 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

730

Methods Data source Data used for this study were retrieved from the health service databases of Lombardy, a industrialized region of Northern Italy that accounts for ∼16% (10 million) of the whole Italian population. In Italy, the population is covered by the NHS, and in Lombardy this has been associated since 1997 with an automated system of databases to collect a variety of information, including (i) an archive of residents who receive NHS assistance (practically the whole resident population), reporting demographic and administrative data; (ii) a public and private hospital discharge database; and (iii) a database on drug prescriptions reimbursable by the NHS. For each patient we linked the information from the different databases via a single identification code. In order to preserve privacy, each identification code was automatically converted to an anonymous code. The reversal of this process was prevented by deletion of the conversion table.

Identifying index events Individuals who in the year 2011 were beneficiaries of the Italian NHS, resident in the Lombardy Region, and hospitalized at least once with a HF diagnosis comprised the study incident cases. Medical records with primary HF diagnosis at discharge were drawn from the regional hospital discharge database. The definitions for a diagnosis for HF included ICD-9 (International Classification of Diseases-9th Revision) codes for heart failure (428.x), and hypertensive heart failure (402.01, 402.11, and 402.91), which were denoted index ICD-9 codes. The inclusion of other ICD-9 codes recently suggested by Saczynski et al.,15 of the Diagnoses Related Group codes for heart failure and shock (127) and of emergency room records concerning HF admission, was also considered. However, we found that 95% of HHF cases were captured from index ICD-9 codes. Thus, with the aim of favouring the most specific criteria for case capture, we preferred the use of our original algorithm including primary index ICD-9 codes from hospital discharge. Because the objective of this study was to look at newly hospitalized HF patients (incident cases), those who experienced multiple hospitalizations contributed only the earliest hospitalization which occurred during 2011. Moreover, patients who had already experienced at least one HHF episode in the 5 years preceding the earliest (or unique) hospitalization occurring in 2011 were also excluded. Incident HHF cases were denoted index events for the current study. They were used for (i) describing their medical history; (ii) measuring incidence rates; (iii) exploring the occurrence of selected outcomes; and (iv) assessing healthcare costs; in other words, to evaluate the burden of patients newly hospitalized for HF in Lombardy.

..................................................................................................................................................................................

the burden of patients newly hospitalized for HF among beneficiaries of the Italian National Health System (NHS) from Lombardy. As reported previously, Lombardy, an Italian region with a population of ∼10 million inhabitants, has a system for linking healthcare utilization databases which is widely used for observational investigations, even in the field of cardiovascular (CV) diseases.14 This system permits analyses of hospitalization, drug prescription, outpatient visits, and mortality on an individual basis. We utilized these linked data to report in detail the incidence, short-term outcomes, and costs of HHF in Lombardy for the year 2011.

G. Corrao et al.

Describing medical history Information about co-morbidities and drug prescriptions occurred during the 5-year period prior the index hospitalization was collected. The corresponding data were drawn from regional archives of hospital discharge, outpatient drug prescription, and drug prescriptions administered directly in day hospital setting, reimbursed by the NHS. Information included history of selected CV therapies, events, and procedures, and respiratory and kidney disease.

Measuring incidence rates Incidence rates of HHF were calculated by dividing the number of index events by the total number of the source population, i.e. beneficiaries of the Italian NHS resident in Lombardy during the year 2011, as recorded by the regional archive of NHS beneficiaries. Rates were crude, stratified for gender and age (10-year categories), and standardized (direct standardization) with respect to the age structure of the European population. Findings were expressed as cases per 10 000 person-years (PYs) at risk. The 95% confidence intervals (CIs) around the estimates were calculated based on the Poisson distribution.

Exploring occurrence of selected outcomes Index events were considered as belonging to the cohort of incident HHF. Cohort members were first evaluated to identify those who died during hospitalization. Among survivors, we identified those who, within 12 months after index hospital discharge, died or were hospitalized at least once. We first estimated the cumulative risks of death and of first hospitalization for any cause. Each cohort member accumulated PYs of follow-up from the date of index discharge until the earliest among the dates of outcome (death or hospital admission) or censoring (death when it is not considered an outcome, emigration, or 12 months after the date of index discharge). Cumulative proportions of patients experiencing the outcome were separately computed by means of the Kaplan–Meier estimator. Secondly, we realized that multiple ‘competing’ outcomes are of interest when hospitalization for any cause has to be split into each cause of hospital admission. We used the competing risks method to estimate the cause-specific cumulative incidence function of hospitalization.16 With this approach, only the first hospital admission which occurred during follow-up was considered, and the overall incidence of admission was decomposed into a sum of the individual cumulative incidence functions for CV and non-CV hospital admission.17 Thirdly, we also realized that the above-described competing risk method considers only the first cause of hospitalization. Under these circumstances, however, causes of hospital admission which occurred after the first one are not taken into account. For this reason, the cumulative proportions of patients experiencing hospitalization for selected causes [i.e. heart failure, major CV events (cerebrovascular or coronary heart events), all other CV events taken together, renal failure, respiratory disease, and/or all other causes of hospital admission] were separately computed by means of the Kaplan–Meier estimator. Another key issue concerns the following question: what would have been the cumulative incidence if patients included in the study cohort had not been hospitalized for HF? We dealt with this question by means of the following procedure. First, a reference cohort suitable to be used as a comparator for the HHF cohort was generated. © 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

731

Reference cohort members were NHS beneficiaries matched 1:1 on gender and age at cohort entry with HHF cohort members, who did not experience hospitalization for HHF and were at risk for the outcome at the time when the matched HHF patient was discharged. Each member of the reference cohort was followed in order to identify those who experienced the considered outcomes. The cumulative incidence of each outcome experienced by HHF and reference cohorts was calculated, and their relative difference was estimated, thus giving a measure of outcome excess due to HHF. The 95% CI around the estimate was calculated based on the normal approximation.

Assessing healthcare cost The total cost for healthcare of each HHF and reference cohort member during the first year after the index hospitalization was measured from the NHS perspective, using the amount that the Regional Health Authority (RHA) reimbursed health providers. Cost categories taken into account were hospitalizations for any cause, outpatient drug prescriptions, and visits, procedures, and diagnostic tests commonly used in the management of CV patients. General practitioner visits were not considered, because they were not recorded in the healthcare utilization databases. Total and per capita healthcare costs (including the index hospitalization) were measured for the HHF cohort. In addition, the difference in cost for healthcare accumulated by every index event and its matched referent (excluding the index hospitalization) was calculated, and the relative difference was estimated, thus giving a measure of the excess healthcare cost due to HHF.

Results Baseline features and medical history Data were initially obtained from 26 949 subjects who, during 2011, had at least one episode of HHF. Among these, 8154 individuals had already been hospitalized within the 5-year period before the index hospitalization. We compared baseline characteristics of these patients with those of 18 795 newly hospitalized patients (Table 1). The average age of newly hospitalized patients was 79 years (SD 11 years), and 51% of them were women (Table 1). At baseline, index patients had a high burden of co-morbid CV and non-CV diseases including hypertension (91%), dyslipidaemia (41%), diabetes (29%), coronary heart disease (24%), AF (18%), stroke (15%), COPD (11%), and chronic kidney disease (9%). More than 50% of the index patients were on treatment with ACE inhibitors (61%), beta-blockers (53%), or diuretics (64%; of whom nearly one-third were receiving aldosterone antagonists). As regards already hospitalized patients, the average age was similar to that of newly hospitalized patients but, as expected, they had a worse co-morbidity and treatment profile.

Incidence rates The crude incidence HHF rate was 29.5/10 000 PYs (95% CI 29.2–29.9). Standardized rates were 31.8/10 000 PYs (95% CI 31.2–32.3) in men and 20.4/10 000 PYs (95% CI 20.1–20.7) in women. The incidence rate increased with age for both genders

........................................................................................................................................................................

Burden of new hospitalization for heart failure

© 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

Table 1 Baseline characteristics and medical history of patients newly hospitalized for heart failure (Lombardy, Italy, 2011) Patients without Patients with previous HHF: previous HHF: total index total index events events (n = 18 795) (n = 8154) ................................................................ Age, mean (SD), years Males Previous hospitalizationsa History of CV disease and proceduresb Hypertension Dyslipidaemia Diabetes Stroke Coronary heart disease Mitral stenosis Atrial fibrillation Atrial flutter Peripheral vascular disease Pacemaker implementation Cardiac defibrillation Use of CV drugsb ACE inhibitors ARBs Beta-blockers Diuretics Aldosterone antagonists Digital glycosides

79 (11) 8988 (47.8%) 7856 (41.8%)

79 (10) 4285 (52.6%) 5542 (68.0%)

17 169 (91.3%) 7672 (40.8%) 5532 (29.4%) 2889 (15.4%) 4535 (24.1%) 1146 (6.1%) 3415 (18.2%) 357 (1.9%) 186 (1.0%) 1388 (7.4%) 213 (1.1%)

8103 (99.4%) 4669 (57.3%) 3433 (42.1%) 1766 (21.7%) 4291 (52.6%) 1901 (23.3%) 4111 (50.4%) 452 (5.5%) 168 (2.1%) 1147 (14.1%) 795 (9.8%)

11 511 (61.3%) 7402 (39.4%) 9900 (52.7%) 12 031 (64.0%) 3570 (19.0%) 2680 (14.3%)

6593 (80.9%) 4108 (50.4%) 6233 (76.4%) 7701 (94.4%) 4998 (61.3%) 2813 (34.5%)

CV, cardiovascular; HHF, hospitalized heart failure. a One year prior to the index hospitalization. b Five years prior the index hospitalization.

(Figure 1). Incidence rates were on average ∼2–3 times higher in men than in women in each age category, except for the oldest (≥80 years) where men had a rate 1.3 higher than that of women (the corresponding values being 614 and 480/10 000 PYs).

Cumulative incidences Of the 18 795 newly hospitalized patients, 1329 (7%) died during the index hospitalization. Of the remaining 17 466 survivors, 4190 died during the following 12 months. Cumulative mortality is shown in Figure 2. Deaths occurred more frequently in the HHF cohort members than in the referent cohort members, cumulative mortalities being 24% and 2.8%, respectively, and the relative difference 88% (95% CI 87–89%). A total of 54 107 hospital admissions and 735 556 days of hospital stays were experienced by the 17 466 patients who survived the index hospitalization. However, ‘only’ 9806 patients were hospitalized at least once during the 1-year period after the index discharge. Cumulative incidences of admission of hospitalized patients are shown in Figure 3. HHF cohort members were readmitted to

732

350 Men

Women

300 250 200 150 100 50 0

≤40

41-50

51-60

61-70

71-80

>80

Age category (years)

Figure 1 Age-specific incidence rates, and 95% confidence interval, according to gender.

0.3

Cumulative incidence

HHF cohort

0.2

0.1 Referent cohort

0 2

4

6

8

10

12

Months from index discharge

Figure 2 Cumulative incidence of out-of-hospital mortality experienced by hospitalized heart failure (HHF) and referent cohort members during the 1-year period after the index discharge (Lombardy, Italy, 2011).

hospital more frequently than referent cohort members, the incidences being 59% and 15%, respectively, and the relative difference 75% (95% CI 74–76%). HHF cohort members were re-hospitalized mainly for CV causes (74.3% of all readmissions), while only 28.6% of the admissions of the referent cohort had a CV cause recorded. Cumulative incidences of cause-specific hospitalizations are shown in Figure 4. With respect to the referent cohort, HHF cohort members experienced a higher incidence of readmission for all the considered causes, the 1-year cumulative incidence being 22% and 0.4% for HF (relative difference 98%, 95% CI 98–99%), 20% and 0.13% for major CV events (relative difference 99%, 95% CI 99–100%), 29% and 5% for all other CV diseases together (relative difference 84%, 95% CI 82–85%), 14% and 1.1% for renal failure (relative difference 92%, 95% CI 91–93%), 23% and 3% for respiratory diseases (relative difference 87%, 95% CI 86–88%), and 17% and 9% for all other causes together (relative difference 43%, 95% CI 40–46%).

........................................................................................................................................................................

Incidence rate/10,000 person-years

G. Corrao et al.

Healthcare costs In the first year after the index discharge, the total direct costs for taking care of the HHF cohort members was €193 million (Table 2). The average cost per person was €11 100, of which €4300 euro was for the index hospitalization (39%), €5900 for the subsequent hospitalizations (53%), and the remaining €900 for non-hospital charges (8%) (Figure 5). By excluding the index hospitalization, the per capita direct cost was €6800 and €1400 for the HHF and the referent cohort members, respectively, the relative difference being 79.4%. The source-specific relative difference was higher for hospitalizations (83.8%) than for outpatient drug prescriptions (59.0%) and services (42.8%).

Discussion This study provides estimates of the incidence, outcomes, and costs of HF in a large, well-defined population from Northern Italy. We found that the incidence rate of newly hospitalized HF was ∼3 cases every 1000 PYs. At least three caveats should be considered for interpreting this estimate. First, our interest is in new hospitalizations rather than ‘de novo’ HF. Secondly, inpatient data probably do not capture all cases of HF, because care is increasingly delivered in the outpatient setting. Thirdly, because of privacy regulations, hospital records were not available, so HF diagnoses cannot be scrutinized and validated. A recent comprehensive study on the misclassification of claims data diagnoses, using medical record review as the gold standard, revealed that the sensitivity of claims diagnoses is often less than moderate, whereas their specificity is usually very good.18 In particular, the specificity for congestive HF from hospital discharge records is expected to be nearly 100% because if a diagnosis is coded and recorded in the claims data it is likely that this diagnosis was made, particularly in hospital discharge summaries.19 On the other hand, not all HF patients are captured from hospital discharge charts, thus explaining the reported 85% sensitivity. By correcting our estimations for these figures, we found standardized incidence rates of 2.4 and 3.8 events per 1000 PYs in women and men, respectively. Differences in (i) demographic and clinical features of the investigated populations; (ii) the data source used for identifying HF cases; and (iii) criteria for heart failure ascertainment, exist across studies. Specifically, incident rates ranging from one to two events per 1000 PYs were reported from studies that used non-validated hospital discharge records from other European countries.20,21 Higher incidence rates were reported in an American study, where the incidence reached a total of six events per 1000 PYs among men and women aged 45–65 years.22 Oddly, these estimates did not differ from those reported from various studies carried out in the USA that used standardized criteria for HF ascertainment, the corresponding rates ranging from two to five events per 1000 PYs according to the Framingham criteria,23,24 or from those reported in a European study based on Boston criteria (3–4 events per 1000 PYs).25 On the other hand, European studies directly capturing cases of incident HF by continuously monitoring participants for the occurrence of HF during follow-up reported higher incidence rates of 17.6 and 12.5 events per 1000 PYs in men and women © 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

733

Burden of new hospitalization for heart failure

Referent cohort

HHF cohort

Cumulative incidence function

0.70

0.20

0.60 0.15

0.50 0.40

0.10 0.30 0.20

0.05

0.10 0.00

0.00 3

6

9

12

3

Time (months since index discharge)

6

9

12

Time (months since index discharge)

Non CV hospital admission

CV hospital admission

aged 55 years or older,26 the corresponding figures in our study being eight and seven events per 1000 PYs, respectively. Other than for total incidence rates, our data are consistent with the increasing rates according to ascending age categories: from a few dozen hospitalized patients per 1000 persons aged ≤40 years, to 20 (women) and 32 (men) events every 1000 persons aged ≥80 years. In contrast, the effect of gender was moderate, i.e. an ∼50% higher rate of hospitalization among men than among women. These estimates are generally similar to those reported from other population-based studies.20 – 26 We found that 7% of newly hospitalized patients died during their index hospital stay. In-hospital mortality from 4% to 10% has been reported from the US ADHERE registry (Acute Decompensed Heart Failure National Registry),27 the Canadian National Mortality Database,28 and the Italian Network on Heart Failure (IN-HF) Outcome Investigation.6 Our results on in-hospital mortality were also similar to those reported in a recent review of studies on post-discharge adverse events among HHF patients, in which in-hospital mortality varied from 3% to 8%.29 As others have done,20 – 23 we documented remarkably high 1-year mortality after HF discharge, the corresponding cumulative incidence being nearly 24%, while mortality expected in the general population was 2.8%, i.e. 88% of the observed mortality among HHF patients was due to HF. Consistent with our results, American and European studies reported 1-year mortality ranging from 21% to 33%.28 – 31 Analogously, a remarkably high 1-year readmission rate after HF discharge was documented in our study, the corresponding cumulative incidence being nearly 59%, while 1-year readmission expected in the general population was 15%, i.e. 75% of the observed readmissions among HHF patients were

.........................................................................................

Figure 3 Overall and cause-specific cumulative incidence of the earliest occurring hospital admission experienced by hospitalized heart failure (HHF) and referent cohort members during the 1-year period after the index discharge (Lombardy, Italy, 2011). Cause-specific hospital admissions are considered as multiple ‘competing’ outcomes, and cause-specific cumulative incidence functions were accordingly estimated by the competing risk method. With this approach, the overall incidence of first admissions after the index discharge was decomposed into the sum of the individual cumulative incidence functions for each cause of hospital admission.

© 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

due to HF. Similarly to our estimates, the ESC-HF Pilot survey, a prospective, multicentre, observational survey conducted in 136 cardiology centres from 12 European countries, reported 1-year mortality and readmission rates of 17% and 44%, respectively.32 Data on the causes of hospital admission show that HF-specific hospitalizations account for only a quarter of all readmissions experienced by HHF cohort members. Analogously, among incident HF cases diagnosed between 1987 and 2006 in the community of Olmsted County, hospitalizations were common after HF diagnosis, but the reason for hospitalization was HF in only 17% of hospital admissions, whereas 62% of hospitalizations were attributed to non-CV causes.33 Conversely, other studies reported that most HHF patients were readmitted due to a CV cause (56–65%) and HF was the most common specific cause.29 These data underscore the major role of co-morbidity in HF34 and that, to reduce the burden of hospital admission in HF patients, strategies must consider both cardiac and non- cardiac conditions. Among the latter, renal failure in particular must be taken into account, due to the very high incidence of hospitalization for this cause (14%) observed in our HHF cohort within 1 year after the index discharge as well as in other studies.35 It should be considered that the incidence of hospitalization for renal failure was very high among patients who did not experience any hospitalization for this cause in the 5 years before the index admission. Finally, we documented that the per capita annual cost borne by the NHS during the first year after the index hospitalization was ∼€11 100, which is very similar to that recently reported by a Sweden study, i.e. an average annual total cost per patient of €11 900.36 The analysis of individual cost items matches the

734

G. Corrao et al.

Cumulative incidence

0.25

0.25

Heart failure

Major cardiovascular events

0.20

0.20

HHF cohort 0.15

0.15

0.10

0.10

HHF cohort

0.05

0.05

Referent cohort

Referent cohort 0.00

0.00 0

3

6

9

0

12

3

Cumulative incidence

0.15

Other cardiovascular events

0.30

12

Renal failure

0.12

HHF cohort

0.25

HHF cohort 0.20

0.09

0.15

0.06

0.10 0.03

Referent cohort

0.05 0.00

0

3

6

9

12

0.00

Referent cohort 0

0.18

Respiratory diseases

0.25

3

6

9

12

Months since index discharge

Months since index discharge

Cumulative incidence

9

Months since index discharge

Months since index discharge 0.35

6

All other causes together

0.15

0.20

HHF cohort

HHF cohort

0.12

0.15 0.09 0.10 0.06

Referent cohort

0.05 0.00 0

3

6

9

12

Referent cohort

0.03 0.00

0

3

6

9

12

Months since index discharge

Months since index discharge

data for other countries,37 indicating that the main cost component is hospitalization (92%), followed by drug prescriptions, medical appointments, laboratory tests, and others, accounting overall for 8% of the total cost. Other authors have reported that the cost of medication represents 11% of the total cost for HF.8,38 This study has several limitations which should be considered. First, as with all administrative database analyses, ICD-9 codes were used to identify diagnoses; these codes may not reflect confirmed clinical diagnoses and lack information to assess the illness severity. Moreover, medical services obtained outside of a patient’s plan are not captured in an administrative database. Pharmacy dispensing may not represent all drugs prescribed or used by the patient (e.g. samples received at a doctor’s office

......................................

Figure 4 Cumulative incidence of hospital admissions for selected causes experienced by hospitalized heart failure (HHF) and referent cohort members during the 1-year period after the index discharge (Lombardy, Italy, 2011).

or hospital, or patients’ out-of-pocket expenses for medications). Thus, resource utilization and costs in our study underestimate the true burden of illness. Finally, an important limitation of our study is that no indirect costs were taken into account. Despite these limitations, our results confirm that the first episode of HF hospitalization is a marker of severe prognosis, since only two-thirds of the patients survived 1 year after admission, and of intensive use of the healthcare system, since only two-fifths of the patients did not experience re-hospitalization 1 year after admission. The costs of diagnostic and therapeutic procedures used in the treatment of HF, from the public payer’s perspective, are therefore very high. With respect to other reports, our study supplies a realistic estimation of the excess of out-of-hospital © 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

735

Burden of new hospitalization for heart failure

Table 2 Total and per capita cost (in thousand of Euros) sustained for healthcare of hospitalized heart failure and referent cohort members during the 1-year period after the index hospitalization, and their relative difference according to source of expenditure (Lombardy, Italy, 2011) HHF cohort Referent cohort Relative difference ............................. ........................... Total Per capita Total Per capita ........................................................................................................................................... Inpatient cost sources Without index hospitalization With index hospitalization Outpatient cost sources Drugs Visits, procedures, and laboratory tests Total Without index hospitalization With index hospitalization

103 211 177 772

5.9 10.2

16 767 –

1.0 –

83.8%

8823 7301

0.5 0.4

3624 4171

0.2 0.2

59.0% 42.8%

119 335 193 896

6.8 11.1

24 563

1.4

79.5%

HHF cohort

Cost source

Referent cohort

In-hospital

Drugs

Visit, procedures and laboratory tests

0

2

4

6

Cost (thousands euro)

Figure 5 Per capita cost sustained for healthcare of hospitalized heart failure (HHF) and referent cohort members during the 1-year period after the index hospitalization (Lombardy, Italy, 2011).

mortality (88%), readmission rate (75%), and healthcare costs (79%) experienced by patients newly hospitalized for HF compared with subjects of the same age and gender. All these data taken together clearly indicate that, despite progress in reducing mortality of patients with chronic HF, hospitalizations for HF remain very frequent39 and represent a relevant clinical and economic burden for both patients and society.

Funding The study was partially supported by an unrestricted grant from Novartis Farma S.p.A. Conflict of interest: none declared.

References 1. Croft JB, Giles WH, Pollard RA, Casper ML, Anda RF, Livengood JR. National trends in the initial hospitalization for heart failure. J Am Geriatr Soc 1997;45:270–275.

.............................................................................................................................................

HHF, hospitalized heart failure.

© 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology

2. Rich MW. Epidemiology, pathophysiology, and etiology of congestive heart failure in older adults. J Am Geriatr Soc 1997;45:968–974. 3. Westert GP, Lagoe RJ, Keskimaki I, Leyland A, Murphy M. An international study of hospital readmissions and related utilization in Europe and the USA. Health Policy 2002;61:269–278. 4. Kozak LJ, DeFrances CJ, Hall MJ. National Hospital Discharge Survey: 2004 annual summary with detailed diagnosis and procedure data. Vital Health Stat 2006;13:1–209. 5. Gheorghiade M, Vaduganathan M, Fonarow GC, Bonow RO. Rehospitalization for heart failure. Problems and perspectives. J Am Coll Cardiol 2013;61:391–403. 6. Tavazzi L, Senni M, Metra M, Gorini M, Cacciatore G, Chinaglia A, Di Lenarda A, Mortara A, Oliva F, Maggioni AP; IN-HF (Italian Network on Heart Failure) Outcome Investigators. Multicenter prospective observational study on acute and chronic heart failure: one-year follow-up results of IN-HF (Italian Network on Heart Failure) outcome registry. Circ Heart Fail 2013;6:473–481. 7. Lloyd-Jones D, Adams RJ, Brown TM, Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, Ferguson TB, Ford E, Furie K, Gillespie C, Go A, Greenlund K, Haase N, Hailpern S, Ho PM, Howard V, Kissela B, Kittner S, Lackland D, Lisabeth L, Marelli A, McDermott MM, Meigs J, Mozaffarian D, Mussolino M, Nichol G, Roger VL, Rosamond W, Sacco R, Sorlie P, Roger VL, Thom T, Wasserthiel-Smoller S, Wong ND, Wylie-Rosett J; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 2010;121:e16–e215. 8. Davis RC, Hobbs FD, Lip GY. ABC of heart failure: history and epidemiology. BMJ 2000;320:39–42. 9. Czech M, Opolski G, Zdrojewski T, Dubiel JS, Wizner B, Bolisega ¸ D, Fedyk-Łukasik M, Grodzicki T. The costs of heart failure in Poland from the public payer’s perspective. Polish programme assessing diagnostic procedures, treatment and costs in patients with heart failure in randomly selected outpatient clinics and hospitals at different levels of care: POLKARD. Kardiol Pol 2013;71:224–232. 10. Braunwald E. Cardiovascular medicine at the turn of the millennium: triumphs, concerns, and opportunities. N Engl J Med 1997;337:1360–1369. 11. Zannad F, Mebazaa A, Juillière Y, Cohen-Solal A, Guize L, Alla F, Rougé P, Blin P, Barlet MH, Paolozzi L, Vincent C, Desnos M, Samii K; EFICA Investigators. Clinical profile, contemporary management and one-year mortality in patients with severe acute heart failure syndromes: the EFICA study. Eur J Heart Fail 2006;8:697–705. 12. Adams KF Jr, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR, Abraham WT, Berkowitz RL, Galvao M, Horton DP; ADHERE Scientific Advisory Committee and Investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE). Am Heart J 2005;149:209–216. 13. Goldberg RJ, Spencer FA, Farmer C, Meyer TE, Pezzella S. Incidence and hospital death rates associated with heart failure: a community-wide perspective. Am J Med 2005;118:728–734.

736

............................................................................................

14. Corrao G, Nicotra F, Parodi A, Zambon A, Heiman F, Merlino L, Fortino I, Cesana G, Mancia G. Cardiovascular protection by initial and subsequent combination of antihypertensive drugs in daily life practice. Hypertension 2011;58:566–572. 15. Saczynski JS, Andrade SE, Harrold LR, Tjia J, Cutrona SL, Dodd KS, Goldberg RJ, Gurwitz JH. A systematic review of validated methods for identifying heart failure using administrative data. Pharmacoepidemiol Drug Saf 2012;21(Suppl 1):129–140. 16. Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999;8:695–706. 17. Kim HT. Cumulative incidence in competing risks data and competing risks regression analysis. Clin Cancer Res 2007;13:559–565. 18. Romano PS, Mark DH. Bias in the coding of hospital discharge data and its implications for quality assessment. Med Care 1994;32:81–90. 19. Kiyota Y, Schneeweiss S, Glynn RJ, Cannuscio CC, Avorn J, Solomon DH. The accuracy of Medicare claims-based diagnosis of acute myocardial infarction: estimating positive predictive value based on review of hospital records. Am Heart J 2004;148:99–104. 20. Jhund PS, MacIntyre K, Simpson CR, Lewsey JD, Stewart S, Redpath A, Chalmers JW, Capewell S, McMurray JJ. Long-term trends in first hospitalization for heart failure and subsequent survival between 1986 and 2003: a population study of 5.1 million people. Circulation 2009;119:515–523. 21. Schaufelberger M, Swedberg K, Köster M, Rosén M, Rosengren A. Decreasing one-year mortality and hospitalization rates for heart failure in Sweden: data from the Swedish Hospital Discharge Registry 1988 to 2000. Eur Heart J 2004;25:300–307. 22. Loehr LR, Rosamond WD, Chang PP, Folsom AR, Chambless LE. Heart failure incidence and survival (from the Atherosclerosis Risk in Community study). Am J Cardiol 2008;101:1016–1022. 23. Levy D, Kenchaiah S, Larson MG, Benjamin EJ, Kupka MJ, Ho KK, Murabito JM, Vasan RS. Long-term trends in the incidence of and survival with heart failure. N Engl J Med 2002;347:1397–402. 24. Goldberg RJ, Spencer FA, Farmer C, Meyer TE, Pezzella S. Incidence and hospital death rates associated with heart failure: a community-wide perspective. Am J Med 2005;118:728–734. 25. Bleumink GS, Knetsch AM, Sturkenboom MC, Straus SM, Hofman A, Deckers JW, Witteman JC, Stricker BH. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure The Rotterdam Study. Eur Heart J 2004;25:1614–619. 26. Remes J, Reunanen A, Aromaa A, Pyörälä K. Incidence of heart failure in Eastern Finland: a population-based surveillance study. Eur Heart J 1992;13:588–593.

G. Corrao et al.

27. Fonarow GC, Adams KF Jr, Abraham WT, Yancy CW, Boscardin WJ; ADHERE Scientific Advisory Committee, Study Group, and Investigators. Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis. JAMA 2005;293:572–580. 28. Tu JV, Nardi L, Fang J, Liu J, Khalid L, Johansen H; Canadian Cardiovascular Outcomes Research Team. National trends in rates of death and hospital admissions related to acute myocardial infarction, heart failure and stroke, 1994–2004. CMAJ 2009;180:E118–E125. 29. Psotka MA, Teerlink JR. Strategies to prevent postdischarge adverse events among hospitalized patients with heart failure. Heart Fail Clin 2013;9:303–320. 30. Harjola VP, Follath F, Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Hochadel M, Komajda M, Lopez-Sendon JL, Ponikowski P, Tavazzi L. Characteristics, outcomes, and predictors of mortality at 3 months and 1 year in patients hospitalized for acute heart failure. Eur J Heart Fail 2010;12:239–248. 31. Jong P, Vowinckel E, Liu PP, Gong Y, Tu JV. Prognosis and determinants of survival in patients newly hospitalized for heart failure: a population-based study. Arch Intern Med 2002;162:1689–1694. 32. Maggioni AP, Dahlström U, Filippatos G, Chioncel O, Crespo Leiro M, Drozdz J, Fruhwald F, Gullestad L, Logeart D, Fabbri G, Urso R, Metra M, Parissis J, Persson H, Ponikowski P, Rauchhaus M, Voors AA, Nielsen OW, Zannad F, Tavazzi L; Heart Failure Association of the European Society of Cardiology (HFA). EURObservational Research Programme: regional differences and 1-year follow-up results of the Heart Failure Pilot Survey (ESC-HF Pilot). Eur J Heart Fail 2013;15:808–817. 33. Dunlay SM, Redfield MM, Weston SA, Therneau TM, Hall Long K, Shah ND, Roger VL. Hospitalizations after heart failure diagnosis: a community perspective. J Am Coll Cardiol 2009;54:1695–1702. 34. Blecker S, Paul M, Taksler G, Ogedegbe G, Katz S. Heart failure-associated hospitalizations in the United States. J Am Coll Cardiol 2013;61:1259–1267. 35. Graziani G, Pini D, Oldani S, Cucchiari D, Podestà MA, Badalamenti S. Renal dysfunction in acute congestive heart failure: a common problem for cardiologists and nephrologists. Heart Fail Rev 2014;in press. 36. Stålhammar J, Stern L, Linder R, Sherman S, Parikh R, Ariely R, Wikström G. Resource utilization and cost of heart failure associated with reduced ejection fraction in Swedish patients. J Med Econom 2012;15:938–946. 37. Berry C, Murdoch DR, McMurray JJV. Economics of chronic heart failure. Eur J Heart Fail 2001;3:283–281. 38. Cline CM, Boman K, Holst M, Erhardt LR; Swedish Society of Cardiology Working Group for Heart Failure. The management of heart failure in Sweden. Eur J Heart Fail 2002;4:373–376. 39. Roger VL. Epidemiology of heart failure. Circ Res 2013;113:646–659.

© 2014 The Authors European Journal of Heart Failure © 2014 European Society of Cardiology